Acta Psychologica 0 North-Holland

47 (1981) 105-115 Publishing Company

SELECTIVE

EFFECTS

ON INFORMATION

H.W. FROWEIN Institute for Perception

Accepted

OF BARBITURATE

PROCESSING

AND AMPHETAMINE

AND RESPONSE

EXECUTION

*

** TNO, Soesterberg,

The Netherlands

April 1980

In a 3 X 2 X 2 factoral experiment, 12 subjects carried out a choice reaction task with reaction time (RT) and movement time (MT) as response measures. Independent variables were drug treatment (amphetamine, barbiturate, placebo), visual stimulus degradation and S-R compatibility. Visual stimulus degradation and S-R compatibility showed additive effect on the RT, but did not affect the MT. This confirms that stimulus encoding, response selection and response execution represent independent processing stages. The two drugs had selective effects on the RT and the MT. Barbiturate (as compared to placebo) had no effect on the MT, but it lengthened the RT, and this effect was additive with the effects of S-R compatibility but showed an interaction with the effects of stimulus degradation. Amphetamine (as compared to placebo) shortened the MT, but there was no significant main effect of amphetamine on the RT although the interaction with the effect of S-R compatibility was significant. These results suggest that barbiturate affects stimulus encoding whereas amphetamine affects response-related processes..

Introduction

This study is part of a project in which the effect of a barbiturate and an amphetamine derivative are investigated in various types of reaction tasks. The strategy of this research consists of trying to identify task variables which are important for the occurrence or size of these drug effects. The idea behind this is that, if these task variables can be related to specific processes or mechanisms which are important in

* Acknowledgements and thanks are due to A.F. Sanders who inspired and supervised this research, to A.J. Krul for carrying out the statistical computations, and to J.Th. Eernst for construction of the apparatus. ** Author’s present address: Centrale Directie der PTT, Afd. CASWO 1, Postbus 30000,250O GA Den Haag, The Netherlands.

105

106

H. W, Frowein /Selective

effects of barbiturate and amphetamine

determining performance, it should also be possible to infer something about the effect of that drug on those processes. When looking for processes that could account for the effects of stimulant and depressant drugs such as amphetamines and barbiturates on performance, it seems obvious to suggest that they affect performance by affecting some sort of arousal mechanism. This idea featured prominently in two previous experiments carried out within this project. In the first experiment, Trumbo and Gaillard (1975) found that barbiturate lengthened the simple RT when the signal consisted of a loud tone, but had no effect when the signal was a small light. Conversely, amphetamine had no effect in the auditory condition but it reduced the RT in the visual condition. To account for this, they suggested that a loud auditory stimulus exerts an “immediately arousing” effect while the visual stimulus would not. It was postulated that barbiturate may act to depress performance by reducing the effect of immediate arousal, while amphetamine could be beneficial when the stimuli themselves are not immediately arousing and it is therefore more difficult to maintain an adequate level of preparation. A similar but more general hypothesis is that depressants such as barbiturates have their greatest effect when the task is somehow arousing, while stimulants such as amphetamines have a greater effect when the task condition is not arousing. This more general hypothesis was also consistent with a subsequent experiment by Frowein and Sanders (1978a) whose findings suggested that barbiturate has a greater effect on RT when the task involves time stress, and that amphetamine is more likely to affect performance when there is no time stress. There are, however, some problems with trying to explain such taskspecific drug effects in terms of changes in arousal. Firstly, it has become increasingly clear that arousal is a more complex phenomenon than thought of at first. Recent activation theories such as those postulated by Broadbent (197 l), Pribram and McGuiness (1975) and Hamilton et al. (1977) all postulate that there must be different types of arousal, but there is little agreement among these different theories when it comes to deciding upon the nature or number of these different arousal types. Hence, the operationalization of specific types of arousal in terms of task conditions remains a somewhat arbitrary matter. Secondly, and perhaps more importantly, arousal theories are concerned with postulating general or specific changes in the state of the organism, but they usually do not specify what this means in terms of the

H. W. Frowein /Selective

effects of barbiturate and amphetamine

107

processes or mechanisms which are necessary to perform a particular type of task. For instance, while Trumbo and Gaillard (1975) proposed that amphetamine may have facilitated the maintenance of arousal when this was not elicited by signals themselves, they suggested that this may have occured either by improving receptor orientation, or by increasing the subjects’ ability to maintain motor preparation or to maintain their attention on the task. An alternative approach to the study of drug effects on performance consists of trying to account for these effects in terms of the component processes which are necessary to carry out the task. A suitable framework for this is provided by Sternberg’s additive factor method. According to the logic of this method, it can be inferred that different task variables affect independent processing stages if they show additive contributions to the reaction time, while an interaction between the effects of different task variables can be assumed to indicate that these variables affect the sa,me processing stage (Sternberg 1969). The additive factor method has been widely used for the identification of processing stages, and one of the most consistent findings has been that the effects of visual stimulus degradation and S-R compatibility are additive (e.g. Sternberg 1969; Shwartz et al. 1977; Sanders 1980). Thus it can be inferred that visual stimulus degradation and S-R compatibility affect two consecutive processing stages which may be called stimulus encoding and response selection. The logic of the additive factor method can also be applied to the relationship between the effects of drugs and the effects of task variables. If a drug and a task variable show an interaction in their effect on the RT, it can be inferred that drug and task variable affect a common processing stage, while additivity implies that they affect separate processing stages. Some examples of an additive factor approach to the study of drug effects are the investigations on marihuana by Darley et al. (1973) and on alcohol by Tharp et al. (1974). The latter study indicated that alcohol consistently impaired response selection but had no effect on stimulus encoding. Also, the additive relations between task variables were unaffected. This indicates that a drug may selectively affect one of several stages while the relation between these stages remains unaffected. In the present experiment, the effects of barbiturate and amphetamine were investigated in a task taken over from a previous experiment by Frowein and Sanders (1978b), in which the subject was presented with a visual signal and had to make a short ballistic movement to one of four targets. In this man-

108

H. W. Frowein /Selective effects of barbiturate and amphetamine

ner the movement time (MT) could be measured as a separate and consecutive measure to the RT. The results obtained by Frowein and Sanders showed that stimulus degradation and S-R compatibility had additive effects on the RT but that they did not affect the MT. Therefore it was concluded that the MT in this type of task represents a stage which is independent of the preceding stages of stimulus encoding and response selection. The purpose of the present experiment was to find out whether either amphetamine or barbiturate have an effect on stimulus encoding, response selection or response execution. The task variables were again visual stimulus degradation and S-R compatibility, and the response measures were reaction time and movement time. Thus, a selective drug effect on visual encoding should be evident from an interaction with degradation, while a drug effect on response selection should result in an interaction with the effect of S-R compatibility on the RT. Similarly, if a drug selectively affects response execution it should show an effect on movement time and not on reaction time; and if a drug affects all three processing stages it should affect both reaction time and movement time and show an interaction with each of the two task variables.

The experiment Method Subjects The Ss were 12 healthy male students from the University of Utrecht ranging in age from 20 to 30 years. Two weeks before participating in the experiment, all Ss received a medical examination and were informed about the nature of the drug treatment and the experimental conditions. They were paid Hfl. 60.00 a day for participating in the experiment and an extra bonus of approximately Hfl. 5.00 to Hfl. 10.00 a day was awarded on the basis of their performance during the experimental task. Drug treatment The treatment conditions consisted of an amphetamine derivative (20 mg Phentermine HCl), a barbiturate (100 mg pentobarbital sodium) and a placebo. Each S received the three treatment conditions on separate days at weekly intervals. Treatment was always administered at 9.00 or 9.30 a.m. by means of a suppository, and the experimental session began 1; hours after treatment and finished about 3; hours later. The pharmacokinetic research by Breimer (1974) and Vree (1973) shows that this should ensure a relatively stable plasma concentration during experi-

H. W. Frowein /Selective

mental sessions. Allocation the Ss nor the experimenter administered. Experimental

effects of barbiturate and amphetamine

109

i.e. neither of the drug treatment was “double-blind”. knew on which days the different treatments would be

(ask and apparatus

The task was a visual four-choice reaction task with reaction time and movement time as the response measures. The S was seated at a sloping desk in a sound attenuating cubicle. The visual signals consisted of flashes generated by a Nixie tube situated about one meter in front of the S. The imperative signal consisted of a 200 msec flash of a diagonal and a horizontal line joining in one of the four corners of the Nixie tube. This stimulus situation is schematically represented in fig. 1. The index finger of the S’s preferred hand rested on the release button, and his task was to make a movement with the index finger to press one of the four target buttons. The distance between the release button and the target button was 13 cm for the two bottom targets and 20 cm for the two top targets. The imperative signal was always preceded by a warning signal, consisting of a 500 msec flash with all elements activated. The inter-stimulus interval between the warning signal and the imperative signal was always one second, and the interval between successive imperative signals was always 5 sec. Ss were specifically instructed that the warning signal served to prepare for a fast response, and that once the movement was initiated it should be made as rapidly and accurately as possible without hesitation about which button to press. They were told that a bonus would be computed on the basis of reaction speed, but that no bonusses would be paid for sessions with more than 3% errors. The

a5

release

Fig. 1. Schematic

representation

button

of the stimulus

situation.

110

H. W. Frowein /Selective

effects of barbiturate and amphetamine

preprogrammed signal presentation and the registration of the responses was performed by the PSARP system (Van Doorne and Sanders 1968). The reaction time (RT) was defined as the interval between the onset of the imperative signal and the release of the release button, and the movement time (MT) was defined as the interval between release of the release button and pressing of the target button. The task variables were S-R compatibility and visual degradation. S-R compatibility was varied as follows: the correct target button in the compatible condition was indicated by the joining point of the two lines (the righthand bottom target in fig. l), while the next target button in counterclockwise direction represented the correct response in the incompatible condition (i.e. the upper righthand button in fig. 1). Visual stimulus degradation was achieved by superimposing a photonegative with a visual noise pattern upon the surface of the Nixie tube. The noise pattern consisted of a cluster of black nonsense shapes, each averaging about 1 mm in diameter. The light-to-dark ratio was about 20%. To avoid differences in light intensity between the two conditions a similar photonegative without noise was used for the undegraded condition. Design

and procedures

Drug treatment (barbiturate, amphetamine, placebo), S-R compatibility and stimulus degradation were varied in a within-subjects design. Thus, each S carried out the reaction task under three treatment conditions and four task conditions. Treatment conditions were varied between days, while S-R compatibility and visual stimulus degradation were varied between sessions but within days. There were 300 trials presented during each session. The order of presentation was varied in the manner of a nested Latin square with the order of SR compatibility conditions nested within the order of treatment conditions, and the order of degradation conditions nested within the order of S-R compatibility conditions. For each S the program consisted of two training days and three experimental days at weekly intervals. During each experimental day, two Ss were alternately tested on each of the four task conditions. Each task condition was tested for a 25min session, and there was a 30-mm rest-period between sessions. Ss were alternately run, so that one S was tested while the other was resting. The first S received the drug treatment at 9.00 a.m. and the first experimental session was at 10.30 a.m., while the program for the second S started 30 min later. Results

The principal measures of performance were the means of RT’s and MT’s These were computed for each individual session and analyzed by separate analyses of variance. The three drug treatment conditions were not analyzed as one variable in these analyses of variance, but separate planned comparisons were made for the effects of barbiturate and amphetamine against placebo. Furthermore, to provide some form of check that the effects of these two drugs and their relationship with the effects of task variables were no artifact of the within-subjects design (Poulton 1973), the data obtained during the first experimental week were separately looked

H. W. Frowein /Selective

effects of barbiturate and amphetamine

111

at. This gives some idea whether different types of results would be obtained with drug treatment as a between-subject variable. Suffice to say that this additional inspection of the data suggests that, if anything, the effects on RT’s and MT’s described below were more prominent with a between-subjects design. The percentages of errors and omissions were also analyzed for each individual session, but very few errors and omissions were made (below 1% for all sessions). Therefore, no additional analyses of variance were carried out on these measures. Reaction

times

Effects of S-R compatibility, stimulus degradation and drug treatment on the mean RT’s are shown in fig. 2. Analysis of variance showed significant main effects of S-R compatibility (F = 443.25; d.f.= 1,24; p < 0.0 1) and visual degradation

wlthout degrod

with degrad

Fig. 2. Reaction time as a function of stimulus degradation, S-R compatibility ment.

and drug treat-

H. W. Frowein /Selective effects of barbiturate and amphetamine

112

(F = 80.71; d.f. = 1,24; p < O.Ol), and there was no evidence of an interaction between these two task variables (F < 1; d.f. = 1,24). Barbiturate-vs.-placebo showed a significant main effect on the RT (F = 12.66; d.f. = 1 ,18; p < 0.0 1) and a small but significant interaction with the effect of stimulus degradation (F = 6.65; d.f: = 1,12; p < 0.05). But the analysis of variance showed no significant evidence of an interaction between the effects of barbiturate and S-R compatibility (F = 2.13; d.f. = 1,12; N.S.) nor of an interaction between the effects of barbiturate, stimulus degradation and S-R compatibility (F = I .66; d.f. = 1 ,12;N.S.). Amphetamine-vs.-placebo, on the other hand, showed a significant interaction with the effect of S-R compatibility (F = 1 1.52; d.j: = 1 ,I 2; p < O.Ol>, but no significant main effect (F < 1; d.f: = 1 ,18). As fig. 2 shows, there was 00 amphetamine effect in the compatible condition, but in the incompatible condition amphetamine resulted in slower RT’s. The other interactions between amphetamine and stimulus degradation (F < 1; d.f. = 1 ,12) and between amphetamine, S-R compatibility and stimulus degradation (F < 1; d.f = 1 ,I 2) were not significant. Movement

times

As is evident from fig. 3, there were no significant effects on the MT of stimulus degradation (F = 2.47; d.f. = 1,24), S-R compatibility (F < 1; d.f. = 1,24) and no interaction between these two variables (F < 1; d.fl = 1,24). The effect of barbiturate VS. placebo was not significant (F = 1.96; d.f. = 1 ,18) and neither were the firstorder interactions of barbiturate with stimulus degradation (F = 1.09; d.f. = 1 ,12) and S-R compatibility (F = 1.49; d.f. = 1,12) or the second-order interaction of barbiturate with stimulus degradation and S-R compatibility (F = 1.OO; d.f. = 1,12). Contrary to this, amphetamine had a significant main effect on the MT (F= 4.72; d.f. = 1,18; p < 0.05); but again there were no significant interactions

;3 -

I

I

1

I

z E

2 ISOlncompatlble

compatible L ILO5

_

5 130-

~=_T’==-=TA-I

0’ E 120 -

-I _____----

.-----•

b

6 al 110E I: , wlthout degrad Fig. 3. Movement

ment,

l

I with degrad

_- A .

I

1 with

without degrgd.

time as a function of stimulus degradation, A- - - - - a = placebo; l A

- - -m = barbiturate;

--¤

degrad

S-R compatibility = amphetamine.

and drug treat-

H. W. Frowein /Selective

between amphetamine and S-R compatibility and S-R compatibility

effects of barbiturate and amphetamine

113

and stimulus degradation (F < 1; d.f, = 1,12), amphetamine (F < 1; d.f. = 1 ,12) or amphetamine, stimulus degradation (F < 1; d.f. = 1 ,12).

Discussion The effects of task variables were consistent with previous findings. The effects of stimulus degradation and S-R compatibility on the RT were additive and neither of these variables had a significant effect on the movement time. Thus, the results confirm that visual encoding, response selection and response execution represent consecutive and independent stages. Moreover, as in the study by Tharp et al. (1974) on the effects of alcohol, this organisation of stages was unaffected by drugs. In this sense, the data provide additional evidence about the robustness of the structural organisation of the reaction process (see also Sanders 1977). Regarding the effects of amphetamine and barbiturate on individual stages, the data clearly show that these drugs have selective effects and that they affect different stages. For amphetamine, the most important effect was the decrease in the MT which indicates that amphetamine speeds up response execution. The data are more ambiguous about the influence of amphetamine on the other stages. The amphetamine X S-R compatibility interaction effect on the RT appears to indicate that for incompatible S-R relations, amphetamine has na inhibitory effect on response selection. But this interpretation is weakened by the failure to obtain a significant main effect of amphetamine on the RT, which would suggest that none of the stages preceding response execution are affected. Thus, although there is clear evidence that amphetamine speeds up response execution rather than encoding or response selection, the evidence with regard to an actual slowing down of the response selection process is more tentative. Barbiturate, unlike amphetamine, did not influence the MT. But there was a significant barbiturate effect on the RT, and this effect was additive with S-R compatibility and showed a small but significant interaction with the effect of stimulus degradation. Thus, the data indicate that barbiturate affects neither response selection nor response execution, but that it tends to slow down the stimulus encoding process. How do these findings relate to the literature about the effects of

114

H. W. Frowein /Selective

effects of barbiturate and amphetamine

these drugs in other types of performance tasks? Regarding the effects of amphetamine, the literature suggests that tasks which mainly involve motor processes are improved, whereas more cognitive tasks are not affected. There is evidence of improved athletic performance (Smith and Beecher 1959), greater grip strength (Hurst et al. 1968) and greater endurance on a bicycle ergometer task (Williams and Thompson 1973); but the review paper by Weiss and Laties (1962) reports that amphetamine had no effect on arithmetic and problem solving tasks or on the Digit Symbol Substitution Test. Similarly, Quarton and Talland (1962) and Talland and Quarton (1965) found no evidence on an effect on the running memory span, and Kopell and Wittner (1968) found that amphetamine had no effect on the identification of forms which were superimposed by visual noise. Thus, the evidence from the literature is consistent with the present finding of selective improvement of the movement time. On the other hand, there is no real evidence in the literature of an inhibitory effect of amphetamine on response selection, and it seems prudent to suggest that this effect will need to be further tested. With regard to the effects of low-dosage barbiturate treatment, the literature shows performance decrements in a great number of different tasks (e.g. Sanders and Bunt 1971). However, because most of these tasks involve both stimulus and response processes, there is little evidence in the literature to corroborate the present indication that barbiturate has a selective effect on stimulus encoding. Nevertheless, there are some recent EEG studies by Otero and Mirsky (1976) and Hink et al. (1978) which indicate that barbiturate depresses the early but not the late components of the evoked potential.

References Berlyne, D.E., 1960. Conflict, arousal and curiosity. New York: McGraw-Hill. Breimer, D.D., 1974. Pharmacokinetics of hypnotic drugs. Doctoral thesis. Nijmegen: Brakkenstein. Broadbent, D.E. 197 1. Decision and stress. London: Academic Press. Darley, C.F., J.R. Tinklenberg, T.E. Hollister and R.C. Atkinson, 1973. Marihuana and retrieval from short-term memory..Psychopharmacologia (Berl.) 19, 231-238. Frowein, H.W. and A.F. Sanders, 1978a. Effects of amphetamine and barbiturate in a serial reaction task under paced and self-paced conditions. Acta Psychologica 42,263-276. Frowein, H.W. and A.F. Sanders, 1978b. Effects of stimulus degradation, S-R compatibility and foreperiod duration on choice reaction time and movement time. The Bulletin of the Psychonomic Society 12, 106-108.

H. W. Frowein /Selective

effects of barbiturate and amphetamine

115

Hamilton, P., B. Hockey and M. Rejman, 1977. ‘The place of the concept of activation in human information processing theory: an integrative approach’. In: S. Dornic (ed.), Attention and performance VI. Hillsdale, NJ: Erlbaum. pp. 463-486. Hink, R.F., W.H. Fenton, J.R. Tinklenberg, A. Pfefferbaum and B.S. Kopell, 1978. Vigilance and human attention under conditions methylphenidate and secobarbital intoxication: an assessment using brain potentials. Psychophysiology 15, 116-124. Hurst, P.M., R. Radlow and Sallyan K. Bagley, 1968. The effects of D-amphetamine and chlordiazepoxide upon strength and estimated strength. Ergonomics 11,47-52. Kopell, B.S. and W.K. Wittner, 1968. The effects of chlorpromazine and methamphetamine on visual signal-from-noise detection. The Journal of Nervous and Mental Disease 147, 418424. Otero, J.B. and A.F. Mirsky, 1976. Influence of secobarbital and chlorpromazine an precentral neuron activity during attentive behavior in monkeys. Psychopharmacologia 46,1-9. Pribram, K.H. and D. McGuiness, 1975. Arousal, activation and effort in the control of attention. Psychological Review 82,116-149. Poulton, E.C., 1973. Unwanted range effects from using within-subject experimental designs. Psychological Bulletin 80, 113-121. Quarton, G.C. and G.A. Talland, 1962. The effects of methamphetamine and pentobarbital on two measures of attention, 1962. Psychopharmacologia 3,66-71. Sanders, A.F., 1977. ‘Structural and functional aspects of the reaction process’. In: S. Dornic (ed.) Attention and performance VI. Hillsdale, NJ: Erlbaum. Sanders, A.F., 1980. ‘Some effects of instructed muscle tension on choice reaction and movement time’. In: S. Nickerson (ed.), Attention and performance VIII. Hillsdale, NJ: Erlbaum. Sanders, A.F. and A.A. Bunt, 1971. Some remarks on the effects of drugs, lack of sleep and loud noise on human performance. Nederlands Tijdschrift voor de Psychologie 26, 670684. Shwartz, S.P., J.R. Pomerantz and H.E. Egeth, 1977. State and process limitations in information processing: an additive factors analysis. Journal of Experimental Psychology: Human Perception and Performance 3,402-4 10. Smith, G.H. and H.K. Beecher, 1959. Amphetamine sulfate and athletic performance I. Objective effects. Journal of the American Medical Association 170,542. Sternberg, S., 1969. ‘On the discovery of processing stages’. In: W.G. Koster (ed.), Attention and performance II. Acta Psychologica 30, 276-315. Talland, G.A. and G.C. Quarton, 1965. The effect of methamphetamine and pentorbarbital on the running memory span. Psychopharmacologia 7,379-382. Tharp Jr., V.K., O.H. Rundell Jr., B.K. Lester and H.L. Williams, 1974. Alcohol and information processing. Psychopharmacologia (Berl.) 40, 33-52. Trumbo, D.A. and A.W.K. Gaillard, 1975. ‘Drugs, time uncertainty, signal modality and reaction time’. In: P.M.A. Rabbit and S. Dornic (eds.), Attention and performance V, 441-454. New York: Academic Press. Van Doome, H. and A.F. Sanders, 1968. PSARP, a programmable signal and response processor. Behavior Research Methods and Instrumentation 1,29-32. Vree, T.B., 1973. Pharmacokinetics and metabolism of amphetamines. Doctoral thesis. Nijmegen: Brakkenstein. Weiss, B. and V.G. Latics, 1962. Enhancement of human performance by caffeine and the amphetamines. Pharmacological Reviews 14, l-36. Williams, M.H. and J. Thompson, 1973. Effect of varient dosages of amphetamine upon endurance. Research Quarterly 44,417-422.